Iterative Deconvolution Method Research Articles

Pre-filtering Effect in an Iterative Deconvolution Method and Its Application to Signal Recovery in Fiber-Optic Distributed Temperature Sensors

The signal quality of a fiber-optic distributed temperature sensor based on optical time domain reflectometry (OTDR) is improved using a pre-filtering technique in the iterative deconvolution method. Investigation of the performance is proposed using iteration conditions in which recovery rate of the signal-component is equal to that of the unfiltering technique. The recovered signal in the pre-filtering technique gives a good signal-to-noise ratio (SNR), 3 times of the unfiltered signal. Consequently, improvements in the OTDR signal with SNR of from 10 to 100 are obtained under the iterations for both distance and temperature resolutions.

Kinetic Studies of Metal Speciation Using Inductively-Coupled Plasma Mass Spectrometry

Kinetic studies of uptake of metal ions by the Chelex batch technique were made to determine Cd, Cu and Pb speciation in model solutions, a snow sample and a river surface water sample. Inductively-coupled plasma mass spectrometry (ICP-MS) and graphite furnace atomic absorption spectrometry (GFAAS) were used for direct determination of these metals. ICP-MS with the solution nebulization technique minimized contamination and adsorption problems involved in the discrete sampling technique of GFAAS, and hence, gave more precise and accurate results. Also, ICP-MS allowed collection of many more data points than GFAAS and was able to resolve components with similar rates of dissociation, which could not always be resolved by GFAAS with its discrete sampling technique. ICP-MS was therefore preferable to GFAAS for kinetic studies of metal speciation. The kinetic data were analyzed by the iterative deconvolution method. The applicability of the Chelex batch technique to metal speciation was validated by analysis of model solutions containing these metal ions with or without EDTA, NTA and fulvic acid. Use of the Chelex batch technique for Cd, Cu and Pb speciation in snow and river surface water samples revealed a number of kinetically distinguishable components of these metals (as complexes) ranging from one to three, probably present as aquo ions or inorganic complexes in the snow sample, and bound to macromolecules/and or colloidal materials in the river surface water sample.

The Shikotan Earthquake of October 4, 1994: Lithospheric earthquake

A large Mw=8.2 earthquake occurred off Shikotan Is., one of the Kurile Is., on October 4, 1994. We inverted 32 body‐wave records to determine the rupture pattern using an iterative deconvolution method. The mechanisms of the subevents were allowed to vary during rupture. The source parameters obtained are: the location of the initial break = (43.48°N, 147.40°E); the centroid depth = 56 km; (strike, dip, rake) = (49°, 75°, 125°) for the total source; the seismic moment Mo = 2.6×1021 Nm (Mw=8.2); source time duration T = 42 s; the average rupture velocity ν = 2.5 km/s. We also determined the mechanism using long‐period Love and Rayleigh waves from 14 stations. The solution for a finite source distributed over a depth range from 0 to 90 km is (strike, dip, rake) = (54°, 76°, 129°) with Mo = 2.3×1021 Nm, in good agreement with that from body waves. Referring to the extent of the aftershock area and the subevent distribution, we estimated the fault area S = 120 × 60 km², the average slip D=5.6m, and the stress drop Δσ=11 MPa. We computed synthetic waveforms as well as static displacements using either the steep or the low‐angle plane as the fault plane, and found that the steep‐dip fault model fits the data better. Our result (the mechanism, large centroid depth, high stress drop) strongly suggests that the 1994 Shikotan earthquake is a lithospheric earthquake: an intra‐plate event that ruptures through a substantial part of the subducting oceanic lithosphere. This type of lithospheric earthquake is relatively common.

Methods and applications for the deconvolution of (Auger) electron spectroscopy data

Spectroscopic data are always the result of a physical process giving rise to some natural line width convoluted with the response function of the respective analyzer. With the aid of numerical deconvolution procedures, however, it is possible to some extent to regain information that was “lost” due to the limited resolution of a given analyzer. A deconvolution is the more necessary the narrower the natural line width of the features compared to the width of the response function (analyzer resolution). In principle, FFT deconvolution with appropriate filtering (e.g. Wiener filtering) is one of the most easiest ways to perform a deconvolution but care has to be taken with sharp features that may lead to oscillatory errors in the deconvolution. Therefore, both mirroring the original spectrum and, in the case of (HRE)ELS data, eliminating the elastic peak is almost essential. The use of an iterative deconvolution method lifts some of these conditions but may produce erroneous structures. Therefore, the reliability of a deconvolution must always be evaluated on the basis of the difference between the original spectrum and the re-convoluted signal.

Kinetic Studies of Metal Speciation Using Chelex Cation Exchange Resin: Application to Cadmium, Copper, and Lead Speciation in River Water and Snow

Kinetic studies of the rate of uptake of the metal ions Cd, Cu, and Pb by the Chelex batch technique have been made on model solutions containing the metal ions and the complexants EDTA, NTA, and fulvic acid. The metal ions were quantitated by inductively coupled plasma mass spectrometry, and the rates of uptake of the metal ions were analyzed by the iterative deconvolution method. In the presence of an excess of EDTA or NTA, all the metal ion was complexed with the chelating agent, forming a slowly dissociating complex. In the presence of fulvic acid, these metals formed strongly bound complexes with a range of slow dissociation rates. The results reveal the importance of [fulvic acid]/[metal ion] ratio and, hence, the extent of occupation of binding sites in fulvic acid (a polyfunctional complexes

Iterative Blind Deconvolution Method with use of Lucy's Algorithm: Application to IR Speckle Data

Lucy's algorithm is applied to iterative blind deconvolution method. This new approach enables to reconstruct a greatly extended object from one speckle frame and to reinforce image domain constraints. A greatly extended object makes speckle images flow over detecting surface. Lucy's algorithm is accommodated to handle such an image and built into iterative blind deconvolution methods. Computer simulation and observational data analyses are conducted and the effectiveness of the proposed methods are exemplified.

Phase diversity imaging through atmospheric turbulence

A method of phase diversity imaging is applied to reconstruct an overly extended object from stellar speckle images. The reconstruction algorithm is based on the iterative blind deconvolution method and the Gerchberg-Saxon phase retrieval procedure. Computer simulations for reconstructing solar granulations are conducted. The simulation results exhibit the effectiveness of the phase diversity method for high-resolution solar imaging.

Parallel scheme of the iterative blind deconvolution method for stellar object reconstruction

An algorithm for stellar object reconstruction without a reference star is described. The algorithm is based on parallel application of the iterative blind deconvolution method to several speckle frames, and it permits reconstruction of an object that satisfies the convolution relation in all the frames used. Computer simulations with noisy speckle images are carried out to demonstrate performances of the algorithm. The algorithm is applied to observational data of binary stars, and high-resolution images are clearly reconstructed. Binary parameters extracted from the images show good consistency with those obtained both with the power spectrum analysis and with the shift-and-add method.

Deconvolution method of atmospheric absorption spectra

The nonlinear iterative deconvolution method is discussed and difficulties in restoring the shapes of spectral lines are noted. Using Newton's iterative method and line-shape theory, an improved method is developed. Numerical experiments are described for the absorption spectra of different line shapes. The results show that our method produces satisfactory deconvolved results for both single- and multiple-line absorption spectra. When pressure broadening is dominant, we find significant improvement for the deconvolved spectra with our procedure.

Extended-object reconstruction with sequential use of the iterative blind deconvolution method

A new algorithm for speckle imaging with sequential use of the iterative blind deconvolution method is described. The algorithm allows reconstruction of an extended object from several noisy speckle frames without reference star data, and calculation of a function to evaluate the reconstructed image. Some results of computer simulations are presented.

Binary Star Reconstruction with Use of the Blind Deconvolution Method

Abstract The iterative blind deconvolution method and a modified auto-correlation method are applied to reconstruct a binary star from a speckle interferogram. The results of two methods are compared to evaluate reconstructed images. For successful reconstruction, a method of frame selection of speckle interferograms is also proposed. A suitable frame selection improves the efficiency of the reconstruction. Some results of the binary star reconstruction are demonstrated to show the usefulness of our method.

A Study on Source Rupture of the 1978 Lan-Hsu, Southeastern Taiwan Earthquake

Seven teleseismic long period P-wave records from World-Wide Standard Seismograph Network (WWSSN) are used to infer the source rupture process of the July 23, 1978 Lanhsu, eastern Taiwan earthquake through the body-wave iterative deconvolution method (Kikuchi and Kanamori, 1982; 1986). Results show that the rupture sequence consists of one main event and one largest subevent near the hypocenter determined by Pezzopane and Wesnousky (1989) and numerous smaller ones at a distance from the hypocenter. Comparison of the spatial distribution of mainshock rupture sequence with aftershocks shows that aftershocks are located in the area which did not break before the occurrence of the above-mentioned events. The total seismic moment of the three largest events is 8.76×10^26 dyne-cm.

Observational Study of a Microburst-Producing Storm Part IV:Fine Scale Analysis

Seven teleseismic long period P-wave records from World-Wide Standard SeismographNetwork(WWSSN)are used to infer the source rupture process of the July 23,1978 Lanhsu,eastern Taiwan earthquake through the body-wave iterativedeconvolution method(Kikuchi and Kanamori,1982;1986).Results show that therupture sequence consists of one main event and one largest subevent near thehypocenter determined by Pezzopane and Wesnousky(1989)and numerous smallerones at a distance from the hypocenter.Comparison of the spatial distribution ofmainshock rupture sequence with aftershocks shows that aftershocks are located in thearea which did not break before the occurrence of the above-mentioned events.Thetotal seismic moment of the three largest events is 8.76×10^(26)dyne-cm.

Iterative blind deconvolution method and its applications

A simple iterative technique has been developed for blind deconvolution of two convolved functions. The method is described, and a number of results obtained from a computational implementation are presented. Some further possible applications are indicated.

Inversion of long-period P-waves from great earthquakes along subduction zones

Long-period P-waves from subduction zone earthquakes in the moment range from 7×10 27 to 8×10 29dyn · cm are analyzed using an iterative deconvolution method. In addition to discrete subevents which effectively contribute to the observed seismograms, the long-period component of the moment rate function is derived with the constraint that the total seismic moment is equal to the moment determined from long-period surface waves. It is shown that great earthquakes can be described by several subevents with a duration of about 10 s. which changes little from earthquake to earthquake in the above moment range. The long-period component of the moment rate function becomes larger for a greater earthquake. The greatest earthquakes of M w ~ 9 additionally involve subevents with another duration scale (much longer than 10 s). It is also shown that the local stress drop is 1 order of magnitude larger than the stress drop averaged over the entire fault plane. The constancy of the local stress drop as well as the average stress drop holds among various magnitudes of earthquakes. These observations suggest an essential similarity in the mode of dislocation distribution.

Determination of Dynamic Forces From Wave Motion Measurements

An experimental method has been developed for generating oblique forces with known orientations and time histories. Recorded signals from several forces were analyzed by an iterative deconvolution method to determine their orientations and time histories. The recovered values agree closely with the exact ones for these controlled sources. These experiments are a valuable test of source characterization methods that may be applied to seismic data from earthquake sources or to signals recorded from the acoustic emission of cracks.

Resolution enhancement in Raman spectra of biological macromolecules by Fourier deconvolution: applications to single-stranded DNA and RNA viruses

The application of a constrained, iterative Fourier deconvolution method to Raman spectra of viruses permits separation of overlapped vibrational bands assigned to viral protein and nucleic acid constituents. The intrinsically broad and extensively overlapped Raman lines, which cannot be resolved by instrumental methods, are sufficiently well separated in the deconvoluted spectra so that band areas may be measured with sufficient accuracy to allow conclusions about the secondary structures and hydrogen bonding interactions of viral molecular components. Deconvolution of Raman scattering in the region 1500–1750 cm −1 of filamentous bacteriophages resolves the contributions of(i)amide I modes ofalpha-helix and beta-strand conformations of the viral coat proteins, (ii) aromatic ring modes of tryptophan, tyrosine and phenylalanine side chains and (iii) purine ring modes of the coated DNA molecule. The results show that among six filamentous viruses the amount of alpha-helix in coat protein subunits decreases in the following order: Pfl (100%) > IKe (93%) > fd (92%) > Ifl (90%) > Pf3 (82%) > Xf (71%). In an application to tobacco mosaic virus (TMV), the complex band in the 800–850 cm −1 region is deconvoluted to resolve contributions from the encapsidated RNA genome at 813 and 822 cm −1, which indicate at least two distinct nucleoside conformers and a contribution from the tyrosine rings of protein subunits at 831 cm −1. Integrated intensity measurements suggest that at least two-thirds of the nucleosides do not contain the usual C3′-endo ring pucker and anti orientation of the glycosidic bond normally associated with nucleoside residues of single-stranded RNA.

Quantitative analysis of nucleic acids, proteins, and viruses by Raman band deconvolution

A constrained, iterative Fourier deconvolution method is employed to enhance the resolution of Raman spectra of biological molecules for quantitative assessment of macromolecular secondary structures and hydrogen isotope exchange kinetics. In an application to the Pf1 filamentous bacterial virus, it is shown that the Raman amide I band contains no component other than that due to alpha-helix, indicating the virtual 100% helicity of coat proteins in the native virion. Comparative analysis of the amide I band of six filamentous phages (fd, If1, IKe, Pf1, Xf, and Pf3), all at the same experimental conditions, indicates that the subunit helix-percentage ranges from a high of 100% in Pf1 to a low of 71% in Xf. Deconvolution of amide I of Pf3 at elevated temperatures, for which an alpha-to-beta transition was previously reported (Thomas, G. J., Jr., and L. A. Day, 1981, Proc. Natl. Acad. Sci. USA., 78:2962-2966), allows quantitative evaluation of the contributions of both alpha-helix and beta-strand conformations to the structure of the thermally perturbed viral coat protein. Weak Raman lines of viral DNA bases and coat protein side chains, which are poorly resolved instrumentally, are also distinguished for all viruses by the deconvolution procedure. Application to the carbon-8 hydrogen isotope exchange reaction of a purine constituent of transfer RNA permits accurate determination of the exchange rate constant, which is in agreement with calculations based upon curve-fitting methods.

Application of an iterative deconvolution method to electron-impact experiments

A procedure based on an iterative deconvolution method is described. An iterative polynomial smoothing associated with the deconvolution method is studied in detail and its importance underlined. Application of the method to dissociative attachment cross-sections yields considerable improvement in the energy resolution of the experimental spectrum. In the case of the C −/CO cross-section the method allows evaluation of the FWHM's of the experimental peaks and leads to a better understanding of the physical processes involved.

Experimental verification of the impulse response method to evaluate transient acoustic fields

An experimental verification of the impulse-response method to evaluate the transient acoustic field of pulsed ultrasonic radiators is presented. Several piezoelectric disk radiators were constructed for the experiments. Transient pressure measurements were obtained at several field points for pulsed wide bandwidth electrical excitations of the radiators. The pressure data were processed via an iterative deconvolution method to obtain experimental impulse responses. The method ideally eliminates the effects of the dynamic response of the transmit and receive system on the impulse responses. A comparison of the experimental and theoretical impulse responses shows the responses to be in general agreement.